Integrating social media and rainfall data to understand the impacts of severe weather in Argentina

Choi, Stella Lina

Permalink

https://hdl.handle.net/2142/90667 Copy

Description

Title

Integrating social media and rainfall data to understand the impacts of severe weather in Argentina

Author(s)

Choi, Stella Lina

Issue Date

2016-04-28

Director of Research (if dissertation) or Advisor (if thesis)

Nesbitt, Stephen W.

Department of Study

Atmospheric Sciences

Discipline

Atmospheric Sciences

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

Argentina, flooding, social media

Abstract

Subtropical South America experiences some of the most intense deep convection in the world. In terms of socio-economic impacts, flooding was the most destructive natural disaster in Argentina between 1980-2010; it affected 343 million people, and caused over $222 billion USD in damages. Furthermore, the national weather service in Argentina has a history of operational issues, the most glaring of which is that a single office in Buenos Aires is responsible for forecasting for the entire country. Consequently, there is a large disconnect between the weather service and the public, which impedes effective severe weather communication. We leverage social media data to understand the public's perception and Twitter activity during heavy rainfall events. Previous studies have investigated the role of social media in circulating critical information during emergencies; however, few have looked into the impact beyond cities in the United States. Rainfall and Twitter activity demonstrate a direct relationship, yet complex, non-linear interactions are likely impacting the results. A new metric, tweeting, efficiency, is developed to account for the inherent lull in social media activity during hours people are most likely asleep. Geo-tagged posts also provide supplemental information, which is particularly advantageous for this region, as it suffers from sparse and biased data.

Graduation Semester

2016-05

Type of Resource

text

Permalink

http://hdl.handle.net/2142/90667

Copyright and License Information

Copyright 2016 Stella Choi

Owning Collections